Isotropic−Cholesteric Phase Equilibrium in Solutions of Cellulose Tris

Mar 29, 2003 - Conio, G.; Bianchi, E.; Ciferri, A.; Tealdi, A.; Aden, M. A. Macromolecules 1983, 16, 1264. [ACS Full Text ACS Full Text ], [CAS]. (4) ...
1 downloads 0 Views 86KB Size
Macromolecules 2003, 36, 2939-2943

2939

Isotropic-Cholesteric Phase Equilibrium in Solutions of Cellulose Tris(phenyl carbamate) Takahiro Sato,*,‡ Toshiyuki Shimizu,† and Fumio Kasabo§ Department of Macromolecular Science, Osaka University, 1-1 Machikaneyama-cho, Toyonaka, Osaka 560-0043, Japan

Akio Teramoto‡ Department of Applied Chemistry, Faculty of Science and Engineering, Ritsumeikan University, Nojihigashi 1-1-1, Kusatsu, Siga 525-8577, Japan Received December 26, 2002; Revised Manuscript Received February 21, 2003

ABSTRACT: The isotropic-cholesteric phase boundary concentrations and interfacial tension were measured for tetrahydrofuran (THF) solutions of fractionated cellulose tris(phenyl carbamate) (CTC) samples. The phase boundary concentrations are compared with the scaled particle theory for the hard wormlike spherocylinder model as favorably as those for other stiffer polymer systems, though the theory does not consider effects of the intramolecular excluded-volume and intermolecular multiple contacts, both of which are appreciable for high molecular weight CTC samples. This may be due to cancellation of the two effects in the isotropic-cholesteric phase equilibrium. The interfacial tension γ between the coexisting isotropic and cholesteric phases for the CTC-THF system slightly decreases with molecular weight, and the asymptotic value is higher than those of other stiffer polymer systems. The dependence of the asymptotic γ on the persistence length is consistent with Cui et al.’s theory based on the second virial approximation, but its magnitude is much larger than the prediction of the theory, indicating that higher virial terms in thermodynamic quantities play a significant role in γ.

1. Introduction It is known that cellulose and its derivatives exhibit lyotropic liquid crystallinity.1-9 They are the most flexible among lyotropic liquid-crystalline polymers, and their chain flexibility shifts the isotropic-liquid crystal phase transformation toward a considerably high concentration region.10 In the previous study (part 1 of this series of papers),11 we have determined the wormlike-chain parameters and intermolecular interaction parameters of a cellulose derivative, cellulose tris(phenyl carbamate) (CTC) in tetrahydrofuran (THF). The chain stiffness of this polymer, characterized by a persistence length q of 10.5 nm, is intermediate between typical stiff and flexible polymers.12,13 It turned out that due to this low chain stiffness, effects of the intramolecular excluded volume and of the intermolecular multiple contacts become important in dilute solution properties of CTC in THF (a good solvent), when the molecular weight M exceeds ca. 105.11 Therefore, CTC exhibits both rodlike and coillike characters by changing M within an accessible M range. The two effects of the intramolecular excluded volume and of the intermolecular multiple contacts may also affect the isotropic-liquid crystal phase equilibrium. So far, the phase equilibria in stiff and semiflexible polymer solutions have been compared with theories,10,14-17 which do not consider the two effects, and good agreements between experiment and theory have † Present address: Asahi Chemical Co. Ltd., Kojima-cho 515, Moriyama, Siga 524-0002, Japan. § Present address: Sumitomo Electric Industry Co. Ltd., Higashisakura 1-1-6, Higashi-ku, Nagoya, Aichi 461-0005, Japan. ‡ Also at CREST of Japan Science and Technology. * Corresponding author: Tel +81-6-6850-5461; Fax +81-6-68505461; e-mail [email protected].

been obtained. Thus, it is an interesting problem to examine whether the theories correctly predict the phase boundary for CTC solutions or not. The isotropic-liquid crystal (cholesteric) phase boundary concentration for cellulose derivatives has been already studied by many workers.1-9 However, most of polymer samples used so far had heterogeneities in the molecular weight and/or degree of substitution, which made the comparison between theory and experiment obscure. In this study, we have used well-fractionated CTC samples with narrow molecular weight distributions and full substitution to measure the isotropiccholesteric phase boundary concentrations of their THF solutions. Since the molecular and interaction parameters of CTC in THF have been already determined in part 1, we can compare experimental and theoretical phase boundary concentrations without ambiguity. In this study, we also measured the interfacial tension between the coexisting isotropic and cholesteric phases of THF solutions of CTC by the sessile drop method. The interfacial tension plays an important role in the kinetics of the isotropic-cholesteric phase separation,18 but there has been no report on the interfacial tension for cellulose and its derivatives solutions so far. 2. Experimental Section CTC Samples. Eleven fractionated CTC samples were used for the phase separation experiment and interfacial tension measurements. Among them, seven samples (F9, F11, F13, F16, F18, F20, F22) were the same as used in part 1,11 and four samples (Q11-4, P2-4, G-4, H-4) were newly prepared in the same procedure as previously. The full substitution of CTC samples was checked in part 1 by elemental analysis. The molecular characteristics of all the samples used in this study are listed in Table 1. For the newly prepared samples, the viscosity-average molecular weights Mv were estimated from

10.1021/ma021786j CCC: $25.00 © 2003 American Chemical Society Published on Web 03/29/2003

2940

Sato et al.

Macromolecules, Vol. 36, No. 8, 2003

Table 1. Molecular Characteristics of CTC Samples Used sample

Mw/104

[η]/102cm3g-1

k′ b

Mw/Mn

F9 F11 F13 F16 Q11-4 P2-4 G-4 F18 H-4 F20 F22

55.5 33.9 15.6 10.3 9.9a 9.6a 7.4a 5.87 4.9a 3.94 2.51

3.35 2.14 1.20 0.834 0.755 0.732 0.568 0.481 0.373 0.339 0.202

0.37 0.38 0.37 0.36 0.39 0.42 0.39 0.41 0.41 0.39 0.45

1.10 1.07 1.08 1.05 1.06 1.05 1.07 1.07 1.06

a Viscosity-average molecular weight estimated from [η] data using the [η]-Mw relation obtained in part 1.11 b Huggins coefficient.

the intrinsic viscosities [η] (in the third column of Table 1) using the relation between [η] and the weight-average molecular weight Mw obtained in part 1. For all the samples, the ratios of Mw to the number-average molecular weight Mn estimated by GPC were equal to or less than 1.1 (see the fifth column of Table 1), which indicates successful molecular weight fractionation of the samples. In the following, Mv is not differentiated from Mw, and both molecular weights are denoted simply by M. Phase Separation Experiment. A biphasic solution was prepared by mixing a CTC sample with distilled THF in a test tube and stirring the mixture by a magnetic chip for 1-2 days in a 25 °C air bath. The solution was then centrifuged at 4000 rpm at 25 °C in a Sorval RC centrifuge to achieve a complete phase separation, and the test tube with the biphasic solution was placed into a thermostated water bath at 25 °C gently without disturbance. The volume of each separated phase was determined from its column height measured by a traveling microscope to calculate the volume fraction ΦLC of the cholesteric phase in the biphasic solution. The test tube had been calibrated to estimate the volume of liquid in the tube from its column height. Measurements of ΦLC were performed by changing the initial CTC mass concentration c and the temperature from 10 to 35 °C. Viscosities of biphasic solutions of CTC solutions with M > 105 were so high that the complete phase separation could not be achieved by centrifugation without disturbing the phase equilibrium. Thus, such solutions were sequentially diluted with THF to find the concentration where the optical birefringence of the solution disappeared using polarizing microscopy under the crossed polarizers condition, and this concentration was taken as the phase boundary concentration cI between the isotropic phase and biphasic regions. The same procedure was used also for the lowest molecular weight sample (F22), whose amount available was too small to make the phase separation experiment. Interfacial Tension Measurements. The isotropic-cholesteric interfacial tension γ was measured for three CTC samples, P2-4, G-4, and H-4, by the sessile drop method.19 After the centrifugation of a biphasic CTC solution at 25 °C, the coexisting isotropic and cholesteric phases were separately taken out by pipets. The separated isotropic phase was put into a glass rectangular cuvette with a 10 mm width, and a suitable amount of the separated cholesteric phase was placed on the bottom of the isotropic phase in the cuvette, using a microsyringe. Evaporation of the solvent THF was prevented by a Teflon cap on the cuvette along with Teflon seal. Pictures of the sessile drop of the cholesteric phase formed on the bottom of the cuvette were taken sequentially at different times t elapsed after forming the drop, under crossed polar conditions by a CCD camera (Hamamatsu Digital Camera 4742-95) with a Micro-Nikkor (55 mm, f/2.8) lens. The magnification of the pictures was determined by a microscope ruler photographed under the same condition (without the cross polares). The sessile drop pictures were analyzed by the established method19,20 to determine γ between coexisting isotropic and cholesteric phases.

Figure 1. Concentration dependence of the density of THF solutions of CTC at 25 °C.

Figure 2. Results of the isotropic-cholesteric phase separation experiments for THF solutions of CTC sample G-4 at different temperatures; c, the initial polymer concentration; ΦLC, the volume fraction of the cholesteric phase in the whole solution. Density Measurements. The sessile drop method needs the difference in the density F between the coexisting two phases to calculate γ, so that F of THF solutions of CTC were measured at 25 °C as a function of the polymer concentration using a Lipkin-Davison-type pycnometer with a 5 cm3 capacity. Figure 1 shows the plot of F against the polymer mass concentration c over a concentration wide range, where c was calculated from the polymer weight fraction multiplied by F. Although the data points at the two highest c are for cholesteric solutions, all data points follow a single straight line indicated. From the slope of this line, the partial specific volume of CTC in 25 °C THF was estimated to be 0.713 cm3/g, being in good agreement with the previous result11 obtained by densitometry at low polymer concentrations (